[1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool

[rav3n_pl]

yeah always nice to wake up to 80% orphan shares....

so what happened the "killswitch" for the older versions take out the new versions too?

http://groups.google.com/group/p2pool-notifications/browse_thread/thread/9d7678823d9931e0
"This problem was triggered by P2Pool's version count going into the >95% upgraded area, everyone switching to the new protocol, and then it briefly going below the threshold, revealing a bug. The root cause of the bug being created was me overlooking the fact that Python's isinstance function also returns True for instances of subclasses. "
Ooops!

[forrestv]

P2Pool is back to normal now. Thanks to everyone who quickly upgraded for generating a forked sharechain that got everyone else mining again.

[Krak]

P2Pool is back to normal now. Thanks to everyone who quickly upgraded for generating a forked sharechain that got everyone else mining again.

So are we back to the pre-update protocol now? Any idea if the luck is fixed?

[Aseras]

ONE does not mention the L word, lol.

On the plus side we hit a block, and someone sent a .5btc donation to the pool.

[rav3n_pl]

P2Pool is back to normal now. Thanks to everyone who quickly upgraded for generating a forked sharechain that got everyone else mining again.

PLZ xplain, what version should be good? I see on todays graph that (blue 2) is coming and going, and (red 1) is always a majority.
We also have VERY unstable global hash rate: 200-500GH/s ? It is still something wrong on network?
Edit:
Also, we have a block found and I not see it on my stats page?

[Gomeler]

Pardon my ignorance on this issue but do I need to change my local p2pool to a different version? Or does the pool just need a core of correct machines to maintain the proper blockchain?


2012-05-18 16:17:57.895323  Local: 5461MH/s in last 10.0 minutes Local dead on arrival: ~1.0% (0-3%) Expected time to share: 8.2 minutes
2012-05-18 16:17:57.896077  Shares: 1223 (523 orphan, 34 dead) Stale rate: ~45.5% (42-49%) Efficiency: ~79.4% (75-84%) Current payout: 0.7746 BTC
2012-05-18 16:17:57.896584  Pool: 416GH/s Stale rate: 31.4% Expected time to block: 5.0 hours

On Thursday I was getting the following
2012-05-17 00:00:01.954858 P2Pool: 17453 shares in chain (17458 verified/17458 total) Peers: 10 (0 incoming)
2012-05-17 00:00:01.954975  Local: 5311MH/s in last 10.0 minutes Local dead on arrival: ~1.5% (0-3%) Expected time to share: 9.4 minutes
2012-05-17 00:00:01.955009  Shares: 525 (48 orphan, 3 dead) Stale rate: ~9.7% (7-13%) Efficiency: ~98.8% (95-102%) Current payout: 0.8951 BTC
2012-05-17 00:00:01.955045  Pool: 334GH/s Stale rate: 8.6% Expected time to block: 6.2 hours


Also, I'm guessing p2pool has had some seriously bad luck in the last week that I've been testing? I'm running around 5 GH/s on p2pool and 4 GH/s on btcguild and my btcguild income has been higher 

[Tittiez]

Also, I'm guessing p2pool has had some seriously bad **** in the last week that I've been testing? I'm running around 5 GH/s on p2pool and 4 GH/s on btcguild and my btcguild income has been higher 

DON'T SAY THE L WORD. The **** has been terrible this week, to answer your question. You can see here: http://p2pool.info/. And yes update your P2Pool, an update was released earlier. Its at the top of the page.

[mdude77]

Also, I'm guessing p2pool has had some seriously bad luck in the last week that I've been testing? I'm running around 5 GH/s on p2pool and 4 GH/s on btcguild and my btcguild income has been higher 

Luck has been AWFUL.

M

[check_status]

90 day luck is consistently 10% below what is possible to be paid out. That would be like getting
⊅45 per block instead of the ⊅50 that is possible within the past 90 days. That isn't just bad luck, but either an attack on coin generation or some unknown design flaw (or a combination of both). I would expect the luck over 90 days to be much closer to 100% than it currently is.

[ModusPwnd]

Why would you expect that?  It really depends on how the luck is calculated and what the luck % actually means right.  Then you can get a confidence interval of how likely it is to be chance.  You cant just guess this kind of stuff though, you have to calculate it.

I guess that 90% means that we are getting 90% of the expectation value.  One thought is that we have so many hashes going on that you would expect real tight statistics, but you have to remember that the probability is exceedingly low so you may not expect the distribution to be as tight as first thought.

Given our number of trials for the last 90 days, and the number of hits is .9 within a std dev.?   

(Im guessing this has probably been discussed elsewhere  )

[Gomeler]

Also, I'm guessing p2pool has had some seriously bad **** in the last week that I've been testing? I'm running around 5 GH/s on p2pool and 4 GH/s on btcguild and my btcguild income has been higher 

DON'T SAY THE L WORD. The **** has been terrible this week, to answer your question. You can see here: http://p2pool.info/. And yes update your P2Pool, an update was released earlier. Its at the top of the page.

Ahh, thanks for that url. I'll update my boxes tomorrow. That explains why more GH/s was making less than a PPS pool. Hopefully the BTC generation increases, I like the idea of p2pool but need to be able to pay for power also..

[check_status]

If luck is the explanation of the frequency of block discovery, then we are discovering 10% below what is possible on average. The below average discovery rate could indicate resource diversion, which has already been shown publicly to occur. You could say that "Luck" is how to explain away resource diversion.

[bitpop]

So you're telling me that every time I get an NMC block, I also found a BTC block? So the 200 NMC I got over the past 2 months, I could have also gotten 200 BTC? I made only like 10 BTC using p2p/pools.

[DeathAndTaxes]

So you're telling me that every time I get an NMC block, I also found a BTC block? So the 200 NMC I got over the past 2 months, I could have also gotten 200 BTC? I made only like 10 BTC using p2p/pools.

No.

NMC difficulty is lower than BTC.  Any block which solves BTC difficulty also solves NMC difficulty (because it is lower).  The reverse isn't true..

[bitpop]

I see, nevermind
Death, what happened to your avatar?

[forrestv]

If luck is the explanation of the frequency of block discovery, then we are discovering 10% below what is possible on average. The below average discovery rate could indicate resource diversion, which has already been shown publicly to occur. You could say that "Luck" is how to explain away resource diversion.

However, P2Pool is open source - there's no closed server that could be secretly diverting hashes. Please, do audit the source. It's either a bug somewhere (which I've been trying like mad to find) or just really bad luck.

[Garr255]

If luck is the explanation of the frequency of block discovery, then we are discovering 10% below what is possible on average. The below average discovery rate could indicate resource diversion, which has already been shown publicly to occur. You could say that "Luck" is how to explain away resource diversion.

However, P2Pool is open source - there's no closed server that could be secretly diverting hashes. Please, do audit the source. It's either a bug somewhere (which I've been trying like mad to find) or just really bad luck.

Regardless, its still worth it to mine with P2Pool!

[kjj]

If luck is the explanation of the frequency of block discovery, then we are discovering 10% below what is possible on average. The below average discovery rate could indicate resource diversion, which has already been shown publicly to occur. You could say that "Luck" is how to explain away resource diversion.

However, P2Pool is open source - there's no closed server that could be secretly diverting hashes. Please, do audit the source. It's either a bug somewhere (which I've been trying like mad to find) or just really bad luck.

Regardless, its still worth it to mine with P2Pool!

+1

10% is totally worth it.  Don't care if it is a bug, a security problem, bad luck, the phase of Ivpeter, "ghosts", whatever.

My post history will show that I started with btc guild in their very early days, and that I have nothing but good things to say about E.  My default distribution of p2pcoin even uses btc guild for backup.  But p2pool is just plain healthier for bitcoin, in the long run.  And I care more about bitcoin today and tomorrow than I care about the balance in my wallet.

[ChanceCoats123]

Agreed. I've been taking the 10% hit like everyone else over the past months. Leaving P2pool and decreasing the hashing power just makes it harder and harder to find blocks and increase our ****. And let's be honest, when the **** is good, it's good. 

[mdude77]

Agreed. I've been taking the 10% hit like everyone else over the past months. Leaving P2pool and decreasing the hashing power just makes it harder and harder to find blocks and increase our ****. And let's be honest, when the **** is good, it's good. 

I was reading through the earlier posts in here and came across this:

http://p2poolparty.org/stats.html

Assuming the link is accurate, there aren't a whole lot of people in p2pool.  I assume some of these addresses are other pools, like p2pmining, but still, the list is smaller than I thought.

M